Commit
•
5cfa89c
1
Parent(s):
76887e4
Update README.md
Browse files
README.md
CHANGED
|
@@ -5,10 +5,59 @@ tags:
|
|
| 5 |
- mergekit
|
| 6 |
base_model:
|
| 7 |
- seyf1elislam/WestKunai-Hermes-7b
|
| 8 |
-
-
|
| 9 |
-
-
|
| 10 |
-
|
| 11 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
---
|
| 13 |
|
| 14 |
# WestKunai-Hermes-10.7b-test
|
|
@@ -70,4 +119,17 @@ pipeline = transformers.pipeline(
|
|
| 70 |
|
| 71 |
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
| 72 |
print(outputs[0]["generated_text"])
|
| 73 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
- mergekit
|
| 6 |
base_model:
|
| 7 |
- seyf1elislam/WestKunai-Hermes-7b
|
| 8 |
+
model-index:
|
| 9 |
+
- name: WestKunai-Hermes-10.7b-test
|
| 10 |
+
results:
|
| 11 |
+
- task:
|
| 12 |
+
type: dataset_type
|
| 13 |
+
name: dataset_split
|
| 14 |
+
dataset:
|
| 15 |
+
name: metric_type
|
| 16 |
+
type: dataset_name
|
| 17 |
+
config: dataset_revision
|
| 18 |
+
split: dataset_args
|
| 19 |
+
revision: metric_name
|
| 20 |
+
args: task_type
|
| 21 |
+
metrics:
|
| 22 |
+
- type: metric_value
|
| 23 |
+
value: dataset_config
|
| 24 |
+
name: task_name
|
| 25 |
+
config: source_name
|
| 26 |
+
args: source_url
|
| 27 |
+
- type: metric_value
|
| 28 |
+
value: dataset_config
|
| 29 |
+
name: task_name
|
| 30 |
+
config: source_name
|
| 31 |
+
args: source_url
|
| 32 |
+
- type: metric_value
|
| 33 |
+
value: dataset_config
|
| 34 |
+
name: task_name
|
| 35 |
+
config: source_name
|
| 36 |
+
args: source_url
|
| 37 |
+
- type: metric_value
|
| 38 |
+
value: dataset_config
|
| 39 |
+
name: task_name
|
| 40 |
+
config: source_name
|
| 41 |
+
args: source_url
|
| 42 |
+
- task:
|
| 43 |
+
type: dataset_type
|
| 44 |
+
name: dataset_split
|
| 45 |
+
dataset:
|
| 46 |
+
name: metric_type
|
| 47 |
+
type: dataset_name
|
| 48 |
+
config: dataset_args
|
| 49 |
+
split: metric_name
|
| 50 |
+
revision: task_type
|
| 51 |
+
args: task_name
|
| 52 |
+
metrics:
|
| 53 |
+
- type: metric_value
|
| 54 |
+
value: dataset_config
|
| 55 |
+
name: source_name
|
| 56 |
+
config: source_url
|
| 57 |
+
- type: metric_value
|
| 58 |
+
value: dataset_config
|
| 59 |
+
name: source_name
|
| 60 |
+
config: source_url
|
| 61 |
---
|
| 62 |
|
| 63 |
# WestKunai-Hermes-10.7b-test
|
|
|
|
| 119 |
|
| 120 |
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
|
| 121 |
print(outputs[0]["generated_text"])
|
| 122 |
+
```
|
| 123 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
| 124 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_seyf1elislam__WestKunai-Hermes-10.7b-test)
|
| 125 |
+
|
| 126 |
+
| Metric |Value|
|
| 127 |
+
|---------------------------------|----:|
|
| 128 |
+
|Avg. |69.75|
|
| 129 |
+
|AI2 Reasoning Challenge (25-Shot)|68.09|
|
| 130 |
+
|HellaSwag (10-Shot) |87.10|
|
| 131 |
+
|MMLU (5-Shot) |64.43|
|
| 132 |
+
|TruthfulQA (0-shot) |64.28|
|
| 133 |
+
|Winogrande (5-shot) |82.72|
|
| 134 |
+
|GSM8k (5-shot) |51.86|
|
| 135 |
+
|